OBJECTIVE The purpose of this study was to systematically compare the effectiveness and safety of full-endoscopic discectomy (FED) and micro-endoscopic discectomy (MED) with open discectomy (OD) for the treatment of symptomatic lumbar disc herniation. METHODS Electronic searches were performed using six databases from their inception to February 2016, identifying all relevant randomized controlled trials and comparative observational studies comparing either FED or MED with OD. Data were extracted and analyzed according to predefined clinical endpoints. RESULTS Twenty three studies were selected for analysis, including 421 FED, 6914 MED, and 21,152 OD cases. No significant difference was found between FED and OD in regards to postoperative visual analog scale (VAS) leg pain scores (WMD 0.03, P=0.93). Similar results were obtained for MED vs OD (WMD 0.09, P=0.18). In terms of postoperative Oswestry disability index (ODI), both FED and MED were similar to OD (WMD -2.60, P=0.32 and WMD -1.00, P=0.21, respectively). FED had a significantly shorter operative duration compared to OD (54.6 vs 102.6min, P=0.0001). MED alone and endoscopic approaches overall (including MED and FED) demonstrated significantly lower estimated blood loss (44.3 vs 194.4mL, P=0.03 and 38.2 vs 203.5mL, respectively, both p<0.05). FED alone demonstrated a trend towards lower estimated blood loss in comparison to OD (3.3 vs 244.9mL, P=0.07). No difference was found in overall complications, recurrence or reoperation rates, dural tears, root injury, wound infections, and spondylodiscitis between FED vs OD, or MED vs OD. CONCLUSIONS Based on this meta-analysis, FED and MED appear to be safe and efficacious alternatives to traditional approaches, but these results require further investigation and validation by prospective randomized studies.